View Item 
      •   IPB Repository
      • Dissertations and Theses
      • Undergraduate Theses
      • UT - School of Data Science, Mathematic and Informatics
      • UT - Statistics and Data Sciences
      • View Item
      •   IPB Repository
      • Dissertations and Theses
      • Undergraduate Theses
      • UT - School of Data Science, Mathematic and Informatics
      • UT - Statistics and Data Sciences
      • View Item
      JavaScript is disabled for your browser. Some features of this site may not work without it.

      Tingkat Kriminalitas dengan Pendekatan Geographically Weighted Random Forest untuk Kota dan Kabupaten di Pulau Jawa

      Thumbnail
      View/Open
      Cover (708.0Kb)
      Fulltext (1.181Mb)
      Lampiran (467.4Kb)
      Date
      2025
      Author
      Dewi, Reyzha Siva
      Djuraidah, Anik
      Silvianti, Pika
      Metadata
      Show full item record
      Abstract
      Heterogenitas spasial tingkat kriminalitas di Pulau Jawa mencerminkan kompleksitas dinamika sosial ekonomi yang tidak seragam antarwilayah. Kondisi ini menimbulkan tantangan dalam memahami faktor-faktor yang memengaruhi tingkat kriminalitas. Oleh karena itu, diperlukan pendekatan analisis yang mampu menangkap variasi lokal pada setiap wilayah. Penelitian ini bertujuan untuk menganalisis keterkaitan peubah sosial ekonomi terhadap tingkat kriminalitas serta mengungkap pola spasial hubungan tersebut. Metode yang digunakan adalah Geographically Weighted Random Forest (GW-RF). Data bersumber dari Badan Pusat Statistik (BPS) tahun 2023 dengan unit analisis berupa kabupaten/kota yang dikelompokkan berdasarkan wilayah hukum Kepolisian Resor (Polres). Peubah penjelas yang dianalisis mencakup tingkat pengangguran terbuka, persentase penduduk miskin, pengeluaran per kapita, Produk Domestik Regional Bruto (PDRB) atas dasar harga berlaku, rata-rata lama sekolah, dan kepadatan penduduk. Proses analisis meliputi pembentukan matriks berbobot spasial, pemodelan dengan GW-RF, serta evaluasi model menggunakan ??2lokal. Hasil penelitian menunjukkan bahwa GW-RF mampu menangkap variasi spasial yang terlihat dari perbedaan nilai ??2lokal antar kabupaten/kota. Secara umum, kepadatan penduduk dan rata-rata lama sekolah merupakan peubah dengan pengaruh terbesar terhadap tingkat kriminalitas. Variasi spasial pada masing-masing peubah memberikan kontribusi yang berbeda pada setiap kabupaten/kota di Pulau Jawa. Temuan ini menegaskan efektivitas GW-RF dalam mengungkap hubungan spasial secara adaptif dibandingkan pendekatan global.
       
      The spatial heterogeneity of crime rates on Java Island reflects the complexity of uneven socioeconomic dynamics between regions. This condition poses challenges in understanding the factors that influence crime rates. Therefore, an analytical approach is needed that can capture local variations in each region. This study aims to analyze the relationship between socioeconomic variables and crime rates and to reveal the spatial patterns of these relationships. The method used is Geographically Weighted Random Forest (GW-RF). The data is sourced from the Central Statistics Agency (BPS) in 2023, with the unit of analysis being districts/cities grouped based on the legal jurisdiction of the Police District (Polres). The explanatory variables analyzed include open unemployment rate, percentage of poor population, per capita expenditure, Regional Domestic Product (RDP) at current prices, average years of schooling, and population density. The analysis process includes the formation of a spatially weighted matrix, modeling with GW-RF, and model evaluation using local ??2. The results of the study indicate that GW-RF is capable of capturing spatial variations as evidenced by differences in local ??2values between districts/cities. In general, population density and average years of schooling are the variables with the greatest influence on crime rates. Spatial variations in each variable contribute differently to each district/city on the island of Java. These findings confirm the effectiveness of GW-RF in revealing spatial relationships adaptively compared to a global approach.
       
      URI
      http://repository.ipb.ac.id/handle/123456789/169259
      Collections
      • UT - Statistics and Data Sciences [82]

      Copyright © 2020 Library of IPB University
      All rights reserved
      Contact Us | Send Feedback
      Indonesia DSpace Group 
      IPB University Scientific Repository
      UIN Syarif Hidayatullah Institutional Repository
      Universitas Jember Digital Repository
        

       

      Browse

      All of IPB RepositoryCollectionsBy Issue DateAuthorsTitlesSubjectsThis CollectionBy Issue DateAuthorsTitlesSubjects

      My Account

      Login

      Application

      google store

      Copyright © 2020 Library of IPB University
      All rights reserved
      Contact Us | Send Feedback
      Indonesia DSpace Group 
      IPB University Scientific Repository
      UIN Syarif Hidayatullah Institutional Repository
      Universitas Jember Digital Repository